maciereader.java

化学图形处理软件

        return entries;
    }
    
    private void createNiceMACiETitle(ChemModel chemModel) {
        chemModel.setProperty(CDKConstants.TITLE,
            "MACIE " + currentEntry.getProperty(EnzymeName) + "= " +
            "PDB: " + currentEntry.getProperty(PDBCode) + ", " +
            "EC: " + currentEntry.getProperty(ECNumber)
        );
    }
    
    private String[] readDtypeDatumTuple(String triggerLine) throws IOException {
        String dTypeLine = triggerLine;
        String datumLine = input.readLine();
        String type = dTypeLine.substring(7);
        String datum = datumLine.substring(7);
        logger.debug("Tuple TYPE: ", type);
        String line = datum;
        if (datum.endsWith("$MFMT")) {
            // deal with MDL mol content
            StringBuffer fullDatum = new StringBuffer();
            do {
                line = input.readLine();
                fullDatum.append(line);
            } while (!(line.equals("M  END")));
            datum = fullDatum.toString();
        } else if (datum.endsWith("+") && (datum.length() >= 74)) {
            // deal with multiline fields
            StringBuffer fullDatum = new StringBuffer();
            fullDatum.append(datum.substring(0,datum.length()-1));
            do {
                line = input.readLine();
                if (line.length() > 0) 
                    fullDatum.append(line.substring(0,line.length()-1));
            } while (line.endsWith("+"));
            datum = fullDatum.toString();
        }
        logger.debug("     DATUM: ", datum);
        String[] tuple = new String[2];
        tuple[0] = type;
        tuple[1] = datum;
        return tuple;
    }
    
    private void processTopLevelField(String field, String datum) 
      throws IOException, CDKException {
        logger.debug("Processing top level field");
        if (field.equals("UNIQUE IDENTIFIER")) {
            currentEntry.setID("MACIE-" + datum);
        } else if (field.equals("EC NUMBER")) {
            currentEntry.setProperty(ECNumber, datum);
        } else if (field.equals("PDB CODE")) {
            currentEntry.setProperty(PDBCode, datum);
        } else if (field.equals("ENZYME NAME")) {
            currentEntry.setProperty(EnzymeName, datum);
        } else {
            logger.warn("Unrecognized ROOT field ", field, 
                        " around line " + input.getLineNumber());
        }
    }
    
    private void processSubLevelField(String field, String fieldNumber,
                                      String subfield, String datum) 
      throws IOException, CDKException {
        logger.debug("Processing sub level field");
        if (field.equals("OVERALL REACTION")) {
            if (subfield.equals("REACTION_ID")) {
                if (readSecondaryFiles.isSet() && readThisEntry) {
                    // parse referenced file
                    String filename = readSecondaryDir.getSetting() + datum + ".rxn";
                    File file = new File(filename);
                    if (file.exists()) {
                        logger.info("Reading overall reaction from: ", filename);
                        FileReader reader = new FileReader(file);
                        IChemObjectReader rxnReader = new ReaderFactory().createReader(reader);
                        currentReaction = (Reaction)rxnReader.read(new Reaction());
                        currentReaction.setID(datum);
                        currentReaction.setProperty(CDKConstants.TITLE, "Overall Reaction");
                        // don't add it now, wait until annotation is parsed
                    } else {
                        String error = "Cannot find secondary file: " + filename;
                        logger.error(error);
                        throw new CDKException(error);
                    }
                } else {
                    logger.warn("Not reading overall reaction for this entry");
                }
            } else if (subfield.equals("OVERALL REACTION ANNOTATION")) {
                parseReactionAnnotation(datum, currentReaction);
                currentReactionStepSet.addReaction(currentReaction);
            }
        } else if (field.equals("REACTION STAGES")) {
            if (subfield.equals("REACTION STAGES")) {
                // new reaction step
                // cannot create one, because CDK io does not
                // allow that (yet)
                reactionStepAnnotation = null;
                reactionStepComments = null;
            } else if (subfield.equals("ANNOTATION")) {
                reactionStepAnnotation = datum;
            } else if (subfield.equals("COMMENTS")) {
                reactionStepComments = datum;
            } else if (subfield.equals("STEP_ID")) {
                // read secondary RXN files?
                if (readSecondaryFiles.isSet() && readThisEntry) {
                    // parse referenced file
                    String filename = readSecondaryDir.getSetting() + datum + ".rxn";
                    File file = new File(filename);
                    if (file.exists()) {
                        logger.info("Reading reaction step from: ", filename);
                        FileReader reader = new FileReader(file);
                        IChemObjectReader rxnReader = new ReaderFactory().createReader(reader);
                        currentReaction = (Reaction)rxnReader.read(new Reaction());
                        currentReaction.setID(datum);
                        currentReaction.setProperty(CDKConstants.TITLE, "Step " + fieldNumber);
                    } else {
                        logger.error("Cannot find secondary file: ", filename);
                    }
                    // convert PseudoAtom's in EnzymeResidueLocator's if appropriate
                    markEnzymeResidueLocatorAtoms(currentReaction);
                    // now parse annotation
                    if (reactionStepAnnotation != null) {
                        parseReactionAnnotation(reactionStepAnnotation, currentReaction);
                    }
                    // and set comments
                    if (reactionStepComments != null) {
                        currentReaction.setProperty(CDKConstants.COMMENT, reactionStepComments);
                    }
                    // now, I'm ready to add reaction
                    currentReactionStepSet.addReaction(currentReaction);
                } else {
                    logger.warn("Not reading reactions of this entry.");
                }
            }
        } else if (field.equals("SUBSTRATES")) {
            logger.warn("Ignoring top level definition of substrates");
        } else if (field.equals("PRODUCTS")) {
            logger.warn("Ignoring top level definition of products");
        } else if (field.equals("REFERENCES")) {
             if (subfield.equals("MEDLINE_ID")) {
                 currentEntry.setProperty(MedlineID, datum);
             }
        } else {
            logger.error("Unrecognized sub level field ", field, 
                        " around line " + input.getLineNumber());
        }
    }
    
    private void markEnzymeResidueLocatorAtoms(Reaction currentReaction) {
    	Iterator containers = ReactionManipulator.getAllAtomContainers(currentReaction).iterator();
    	while (containers.hasNext()) {
    		IAtomContainer ac = (IAtomContainer) containers.next();
    		for (int i=0; i<ac.getAtomCount(); i++) {
    			IAtom atom = ac.getAtom(i);
    			if (atom instanceof EnzymeResidueLocator) {
    				// skip atom
    			} else if (atom instanceof PseudoAtom) {
    				PseudoAtom pseudo = (PseudoAtom)atom;
    				logger.debug("pseudo atom label: ", pseudo.getLabel());
    				logger.debug("pseudo class: ", pseudo.getClass().getName());
    				Matcher residueLocatorMatcher = residueLocator.matcher(pseudo.getLabel());
    				if (residueLocatorMatcher.matches()) {
    					logger.debug("Found residueLocator: ", pseudo.getLabel());
    					// replace atom with enzymeResidueLocator
    					IAtomContainer container = ReactionManipulator.getRelevantAtomContainer(
    							currentReaction, pseudo
    					);
    					logger.debug("Replacing the pseudo atom with a ezymeResidueLocator atom");
    					AtomContainerManipulator.replaceAtomByAtom(container, 
    							pseudo, new EnzymeResidueLocator(pseudo));
    				}
    			}
    		}
    	}
    }
    
    private void parseReactionAnnotation(String annotation, Reaction reaction) {
        logger.debug("Parsing annotation...");
        Matcher annotationTupleMatcher =
            annotationTuple.matcher(annotation);
        while (annotationTupleMatcher.matches()) {
            String field = annotationTupleMatcher.group(1);
            String value = annotationTupleMatcher.group(2);
            processAnnotation(field, value, reaction);
            // eat next part of annotation
            String remainder = annotationTupleMatcher.group(3);
            annotationTupleMatcher =
                annotationTuple.matcher(remainder);
        }
    }

    private void processAnnotation(String field, String value, Reaction reaction) {
        logger.debug("Annote: ", field, "=", value);
        if (field.equals("RxnAtts") || field.equals("RxnType")) {
            // reaction attributes
            /*String dictionary = "macie";
            if (value.equals("Acid") || value.equals("Base")) {
                dictionary = "chemical";
            }*/
            addDictRefedAnnotation(reaction, "Attributes", value);
        } else if (field.equals("ResiduesPresent") ||
                   field.equals("GroupTransferred") ||
                   field.equals("BondFormed") ||
                   field.equals("ReactiveCentres") ||
                   field.equals("BondCleaved") ||
                   field.equals("BondFormed") ||
                   field.equals("Products") ||
                   field.equals("ResiduesPresent")) {
            reaction.setProperty(new DictRef("macie:" + field, value), value);
        } else if (field.equals("Reversible")) {
            if (value.equalsIgnoreCase("yes")) {
                reaction.setDirection(Reaction.BIDIRECTIONAL);
                addDictRefedAnnotation(reaction, "ReactionType", "ReversibleReaction");
            }
        } else if (field.equals("OverallReactionType")) {
            StringTokenizer tokenizer = new StringTokenizer(value, ",");
            int i = 0;
            while (tokenizer.hasMoreTokens()) {
                String token = tokenizer.nextToken();
                i++;
                reaction.setProperty(
                    DictionaryDatabase.DICTREFPROPERTYNAME + ":field:overallReactionType:" + i, 
                    "macie:" + token.toLowerCase()
                );
            }
        } else {
            Matcher residueLocatorMatcher =
                residueLocator.matcher(field);
            if (residueLocatorMatcher.matches()) {
                logger.debug("Found residueLocator: ", field);
            	boolean found = false;
                Iterator containers = ReactionManipulator.getAllAtomContainers(reaction).iterator();
                while (containers.hasNext()) {
                	IAtomContainer ac = (IAtomContainer) containers.next();
                	logger.debug("Searching for given residueLocator through #atom: ", ac.getAtomCount());
                	logger.debug("Taken from reaction ", reaction.getID());
                	for (int i=0; (i<ac.getAtomCount() && !found); i++) {
                		if (ac.getAtom(i) instanceof PseudoAtom) {
                			// that is what we are looking for
                			PseudoAtom atom = (PseudoAtom)ac.getAtom(i);
                			if (atom.getLabel().equals(field)) {
                				// we have a hit, now mark Atom with dict refs
                				addDictRefedAnnotation(atom, "ResidueRole", value);
                				found = true;
                			}
                		}
                	}
                }
                if (!found) {
                    logger.error("MACiE annotation mentions a residue that does not exist: " + field);
                }
            } else {
                logger.error("Did not parse annotation: ", field);
            }
        }
    }
    
    private void addDictRefedAnnotation(IChemObject object, String type, String values) {
        StringTokenizer tokenizer = new StringTokenizer(values, ",");
        while (tokenizer.hasMoreTokens()) {
            String token = tokenizer.nextToken();
            object.setProperty(new DictRef("macie:" + type, token), token);
            logger.debug("Added dict ref ", token, " to ", object.getClass().getName());
        }
    }
    
    public void close() throws IOException {
        input.close();
    }

    private void initIOSettings() {
        selectedEntry = new IntegerIOSetting("SelectedEntry", IOSetting.LOW,
          "Which entry should I read?",
          "1");

        readSecondaryFiles = new BooleanIOSetting("ReadSecondaryFiles", IOSetting.LOW,
          "Should I read the secondary files (if available)?",
          "true");

        readSecondaryDir = new StringIOSetting("ReadSecondaryDir", IOSetting.LOW,
          "Where can the secondary files be found?",
          System.getProperty("user.home") + System.getProperty("file.separator"));
    }
    
    private void customizeJob() {
        fireIOSettingQuestion(selectedEntry);
        fireIOSettingQuestion(readSecondaryFiles);
        fireIOSettingQuestion(readSecondaryDir);
    }

    public IOSetting[] getIOSettings() {
        IOSetting[] settings = new IOSetting[3];
        settings[0] = selectedEntry;
        settings[1] = readSecondaryFiles;
        settings[2] = readSecondaryDir;
        return settings;
    }
    
}